DESCRIPTION Hemolytic uremic syndrome (HUS) is an important cause of acute renal failure (ARF) in children. The most common form of this disease, D+HUS, is associated with exposure to enterohemorrhagic strains of E. coli. These bacteria, both 0157:H7 and non-0157:H7 serotypes, elaborate Shiga-like toxins (SLT) and cause hemorrhagic colitis. The SLT provokes diffuse injury to endothelial cells and a systemic vasculopathy which leads to erythrocyte and platelet destruction and kidney damage. Nearly 15-20% of patients with D+HUS develop severe ARF and require dialysis support. The acute mortality rate is 5-10% and serious extra-renal complications occur in 15-20% of patients with D+HUS. There is currently no specific therapy for D+HUS. All previous interventions have been aimed at reversing the pathophysiological events which follow dissemination of SLT. Therapy designed to limit SLT absorption from the gastrointestinal tract before there is systemic exposure to bacterial toxin and cytokine activation might lower the mortality rate, ameliorate ARF, and reduce the severity of organ injury in D+HUS. SYNSORB Pk is an adsorbent agent composed of a platform molecule, diatomaceous silicon dioxide, which is covalently linked to a unique oligosaccharide chain. SYNSORB Pk irreversibly binds enterohemorrhagic E. coli associated SLT. This agent is free of significant toxicity. We propose to test the hypothesis that early treatment with SYNSORB Pk will blunt absorption of SLT, thus reducing the occurrence of catastrophic complications, i.e., mortality and severe extra-renal events, and lower the frequency of ARF requiring dialysis. The validity of this assertion will be tested in a randomized, double-blind clinical trial comparing the effect of SYNSORB, Pk with that of placebo in children with newly diagnosed D+HUS. Patients will be recruited from member institutions of the Council of Pediatric Nephrology of the National Kidney Foundation New York/New Jersey. The effect of the therapy on patient mortality, serious extra-renal complications, and the need for dialysis will be assessed. The results of treatment will be correlated with the microbiological etiology of D+HUS, the duration of stool excretion of free SLT, and the degree of cytokine activation in the stool and plasma. The outcome of this trial may justify the use of SYNSORB PK as a safe therapy to ameliorate D+HUS.